Hi Roger,

Just guessing, but probably your "simple" select ending up using an
Index.
Maybe q filespace * * doesn't.

Try coding a select that doesn't use the index, see if you can slow it
down! :>)



-----Original Message-----
From: ADSM: Dist Stor Manager [mailto:ADSM-L AT VM.MARIST DOT EDU] On Behalf Of
Roger Deschner
Sent: Monday, January 30, 2006 4:00 PM
To: ADSM-L AT VM.MARIST DOT EDU
Subject: Why is Q FILESPACE so slow?


The TSM QUERY FILESPACE command, with a wildcard for nodename, takes
FOREVER. On our system, a big TSM server with about 1000 nodes, a Q
FILESPACE * * can take 20 minutes to an hour, depending on how busy the
server is doing other things.

I finally got tired of this sluggishness, and put together a simple SQL
SELECT that asked the same question, just to see what would happen. It
was fast.

What is going on here? Why should the performance difference be about
two orders of magnitude? (100x) (As I recode a bunch of scripts I have
to change Q FILESPACE into SELECT.)

Roger Deschner      University of Illinois at Chicago     rogerd AT uic DOT edu
============= What if there were no rhetorical questions? ==============